Content area
Abstract
Various types of noise and nonlinearities in digital magnetic recording systems are investigated in this dissertation. Measurement techniques and analyzing methods are developed to understand each phenomenon. The nonlinearities due to the replay process using MR sensors are studied in Chapter 4. The nonlinearities are determined by comparing the measured signal with that obtained from a linear analysis. A characterization method of transition noise is developed in Chapter 5. Approximating transition noise by several leading "modes" allows the noise parameters to be determined experimentally. Chapter 6 covers the investigation of disk substrate texture induced noise. The noise mechanism and characteristics are systematically studied. An analytical noise correlation function that directly relates the noise with the fluctuations of the textured disk surface is also developed in this chapter. An error rate model including colored and nonstationary noise is developed to further understand the impact of noise on system performance in Chapter 7. Noise with different characteristics is shown to influence the system performance differently. In addition, the influence of texture noise is examined in term of each noise parameter based upon the noise model developed in Chapter 6. Finally, in Chapter 8, the effect of finite write field rise time on recording performance is studied. Recording performance predicted by a simplified analytical model is compared with the measurements. It is shown that a slow flux rise time causes a degraded field gradient during writing, which results in a broader written transition, a larger NLTS, and noisier transition boundaries.
